CA1328303C

CA1328303C - Position controller for glass sheet processing system

Info

Publication number: CA1328303C
Application number: CA000534921A
Authority: CA
Inventors: Gary D. Brinker; Jeffrey N. Klopping; Eric D. Fintel
Original assignee: Glasstech Inc
Current assignee: Glasstech Inc
Priority date: 1986-04-17
Filing date: 1987-04-16
Publication date: 1994-04-05
Anticipated expiration: 2011-04-05
Also published as: JPH02500104A; EP0302078A4; EP0302078A1; BR8707668A; US4782449A; FI884761A; FI884761A0; AU7358087A; ATE107422T1; FI94561B; DE3750092D1; KR950012517B1; KR880701413A; AU614261B2; WO1987006373A1; FI94561C; EP0302078B1; DK604587D0; DE3750092T2

Abstract

ABSTRACT
A position controller for a glass sheet processing system having a central control system including an operator interface and a master computer, the position controller including a slave computer, a port for two-way communication of multi-character command and data signals between the master computer and the slave computer, a digital to analog converter, a variable speed drive for driving a movable component in the glass processing system, a port for communicating a signal from the slave computer through the digital to analog converter for driving the variable speed drive, and absolute position encoder associated with the movable component in the system, and a port for communicating the digital signal representative of the position of the movable component to the slave computer. The slave computer system includes logic for receiving data and command input from the master computer in the central control system, processing this information as required, generating a signal capable of operating a vehicle speed drive, and reporting an echo back signal to the master computer indicating that the driven component has reached the desired position. The slave computer also includes logic for transmitting current position and stored end point and velocity profile information back to the master computer for output to the monitor at the operator's request.

Description

POSITION CONTROLLER FOR GLASS :~
S~EE~ PROCESSING SYS~EM

TECB~IICAL FIEL~

This invention re~ate~ generally to control~ for large g~ass processing systems, and more particularly to a modular po~ition controller and communication interface system u~ed in connec-tion with the central control of a qlass processing systom. ~;

BACRGROUND ART
., " . .
Glass sheet proce6sing systems such as the type disclosed by U.S. Patent No. 4,575,390 include bending app~ratus having one or more molds adapted to be positioned within a heating chamber and receive a heated glass Jheet from a roller ;~-conveyor in preparation for tompering and/or b-ndlng. -Briefly, the gla~s processing system typically includes a furnace defining a hoating :~
cha~ber through which gl~ss sheet~ are conveyed for heatlng in preparatlon for bending. ~he bending apparatu- of the preferred system lncludes a rollor conveyor for supplying heated glass to one or more curved moldfi. ~he curved molds typically take the shape of a ~urface having a complex curvature that is generally convex in nature or a complimentary concave surface in the form of an open center ring.
The heated gla~s is formed by placing the 6heet in ~ ' .', ': .' a series of steps on a mold and moving the mold(s) relative to the glass to provide an accurately formed curvature according to a preselected design.
The molds are each typically mounted for movement along a single axis. Thus, a position controller of the present invention would be required for each of these movable members.
The qulck and accurate positioning of the mold6 during various stages of the process is an important factor in achieving a high quality product ln this bending and tempering process.
Thu~, the processing system must include a central control system capable of simultaneously monitoring various conditions throughout the syste~ and simultaneously positioning variou~ movable compo-nents of the system according to the process.
The central control system typically includes an operator interface or console which may be in the form of a teletype unit for inputtlng variou~ data, such a- ~elected important mold po~tions and deslred temperatures, lnto a master computer. ~he master computer monitors variou~
conditions, such as the actu~l temperature at variou,~ ~elected points in tho furnace, and trans-mit~ this information to the operstor through theconsole or other suitable data output device.
The ma~ter computer al~o communicates with one or more position controllers. ~he posi-tion controller processes positioning co~mand~
received from the ma~ter computer, receives input from a position sensor located on the driven component, and issues a signal to activate the - :

P~422 -3-variable speed drive unit for that component cau~in~ the driven component to move according to the appropriate velocity profile to the desired point.
One disadvantage of existing glass processinq position control systems is that commu-nication between the master computer and the slave computer is limited to two-wire open-loop transmis-sion of single ASCII character commands. This limited communication, while sufficient to allow for transmis~ion by the master computer of single character motion command~, and transmission by the slave computer of single character ac~nowledgment commands, greatly restricts the flexibility of the system.
For example, the master computer cannot receive actual position information for any compo-nentC from the slave computer~ Thus, if the operator has manually moved (~ogged) any of the component~ to a different position, the slave controller can ascertain this new position via its communicatlon wlth a posit~on encoder, but the master computer i8 not updated accordingly.
Similarly, if it i8 desirable that a component move to a selected end point and then oscillate hetween two points from this position for a selected period of time, the master computer does not track the location of the mold at all times during this oCcillation routine.
Also, as a result of the limited communi-cation capabilities between the ma~ter and slave controller, the positions and desired velocity ,. : :. : . . :~ . . . ~ : :, ... ; - , . . : - .: : . .

profiles for the end points cannot be downloaded by the master computer. The locations of preselected points (nend points~), along with the drive parame-ters and move characteristics for those points are permanently burned" into an Electrically Program-mable Read-Only Memory (EPRON) and cannot be downloaded or otherwise changed from the master computer.
Another disadvantage of existing systems is that, in situations where a movable component i~
periodically shuttled into and out of a furnace during the glass processing cycle, the shuttle upon which the component is mo~nted often undergoes thermal expansion or contraction. Thus, the actual location of the component changes during process-ing. This uncontrollable thermal expansion/con-traction causes positioning problems, particularly when the component is programmed to move to a ~elected end point where accurate positioninq ifi important, such as where one mold is to be mated with another mold. The operator i6 thus forced to make any compensation for this thermal expansion or contraction on the basis of his observation of the change in position of the mold.
Another disadvantage of present control system6 is that the master computer board is different in configuration from the slave control-ler board 80 that separate replacement boards for each of the controllers needs to be kept in stock.
Also, though the slave controller boards for each of the different movable molds i8 identical, separate replacement ROM chips, each corresponding - .. . . . . .~ ,; ~ . ~ - ., :,. . - . . ~ . . :. . . . :

.: , . ,: . - .

1 328303 ::

to a particular mold or other movable component, must be kept in stock.
: .
SUMMARY OF THE INVEN~ION
- - .
One object of the present invention is to ~--provide a position controller for controlling and monitoring positionable components, such as mold&, in a glas6 sheet processinq system.
Another object of the present invention is to provide a position controller including a slave computer capable of receiving and stor1ng data corresponding to preselected end points and velocity profile data for each of those end points on the glass processing system, wherein said data~
downloaded from the master computer in the central control system at setup time or at any time during operation of the system.
A further object of the present invention is to provide a position controller including a ~lave computer having an automatic thermal expan-sion compensator which, in response to changes in the length of the shuttle due to thermal expansion or contraction, automatically alters the value of the end point po~ition for selected points to ensure that a movable component is correctly positioned aespite the change in length of the shuttle upon which the component is mounted.
It is yet another object of the present lnvention to provide a position controller includ-ing a slave computer having communication means for transmitting data, such as current position information, or the value and/or the velocity profile for a selected end point, at any time in response to the operator'~ request transmitted from the master computer.
Another object of the present invention is to provide a position controller including a slave computer having means for receiving po~ition command~ or data from an input source such as an operator terminal or a suitably programmed master computer, sortlng the commands and data according to a predefined hierarchy, and processing those commands and/or data in the order corresponding to the predefined hierarchy rather than in the order that the commands and/or data are received.
A further object of the present invention i8 to provide a position controller including a slave computer having means for determining when a movable component has made an unprogrammed stop durinq a programmed motion, and mean~ for altering the velocity profile for that component in response to the unexpected ~toppage.
Another object of the present invention i8 to provide a position controller including a slave computer having the capability to detect an abrupt commanded change in dlrection or rapid deceleration of the molds and provide controlled, time-based acceleration of the molds in order to achieve a smooth change in speed and/or direction.
The position controller of the present invention i8 adapted for use with a ma~ter computer having supervisory control and monitoring of the various conditions in the glass sheet processing system, and includes a slave computer, in the form of a programmable microprocessor, first input means capable of receiving posltion command~ and position data from the master computer, logic means for processing this information as required, and flrst output means for generating an analog signal capable of operating a variable speed drive. The slave computer also includes second input means for receiving digltal signals from a position encoder located at a fixed point on the furnace for monitoring the current position of a mold or other movable component, and ~econd output means for transmitting current position information back to the master computer where it iB
output to the monitor at the operator's request.
The slave computer is programmed to receive positioning commands, retrieve end point values and velocity profiles associated with the po~ltioning command received from the master computer and generate a series of signals whlch actlvate a variable speed drlve, thereby causing motion of the controlled axi6 to the desired polnt wlth the desired start-up acceleratlon, traverse velocity, and deceleratlon for the reque~ted motlon. The parameters utllized to develop the requislte informatlon for po~itlonlng to any of the ~elected polnts, lncludlng the aosoclated veloclty proflles for tho~e polnts, may be downloaded to the slave computer from the master computer at ~etup.
Communlcation~ protocol, allowlng for two-way tran~ml~ion of character ~trlng~ up to 80 characters long ~etween the master computer and slave computer, allows end point po~itlon and '. .

': ' : ' '' . ., ' ` : `: i " ' '' ' ;' : ' ' 7 ~, 1 3283~3 velocity profile data to be changed and downloaded at any time.
Data is received by the slave computer from a position ~o~oer on the moving components of the system via multibit parallel data lines. This information is utilized by the slave computer system to monitor the current position of the movable component~ on the furnace.
The silave computer also include~ a programmable interval timer which, in conjunction with current position information, is utilized to determine an unexpected "STOP" condition and, in response, generate the necessary signals to auto-matically alter the current velocity profile to achieve a controlled time-based restart and posi-tioning of the movable component to the desired end point from the unexpected stop position.
The slave computer syitem also include~
means for determining if any programmed velocity would cause ~ change in the speed of the drive greater than a predetermined threshold. The slave computer sy~tem then automatically adjusts the velocity signals output to the drive to ensure a smoother transition in ~peed and/or direction.
The slave computer system includes an Automatic Thermal Expansion Compensator (ATEC).
The ATEC feature automatically modifieei the posi-tion of a preselected end point in re~ponse to a detected change in the physical length of the shUttle caused by thermal expansion or contraction of the shuttle during heating or cooling as the shuttle is moved from the inside to the out~ide of the furnace.
A software driven ~jog~ function i~ also provided in the slave computer system which is activated through the communications link with the master computer. This function allows the operator to employ an electric jog switch from the console to move the movable component in either direction for a period of time correspondinq to the opera-tor's activation of the switch.
Similarly, the jog switch may also beemployed, in conjunction with software control of the slave computer system, to position the movable component by a single incremental unit in either direction for more accurate manual positioning.
~ue to the increaced communication capabilities of thi~ system, these jog and plus/minus motion functions allow the operator to manùally po3ition the movable components during setup time, ascertain the actual position readings by interrogating the ~lave computer, and deflne these positions as end po$nts for use during the process.
The use of a slave computer to ~enerate the specific signals necessary to operate the vari-able speed drive for a particular driven component,based upon general position commands received from the master computer and actual position information received from a sensor located on the machine, simplifies and generalizes the tasks of the master computer and increases modularity within the entire control system. This increa~ed modularity allows :'.' , , ', ,' ' , ., ;, .: ~ ' , . . . X- ,'. -' ".

,. ' ' '' ', . ':. ' , . ., ., . ' .',, ' .'~' '" '.. ; ' ' ' , ' ",: .': ' ' , '':

1 3283~3 for easier maintenance and replacement of individu-al components in the overall control sy~tem.
Al~o, the various components of the control ~ystem, including the operator console, the master computer, and the slave computer, can be located remotely from the furnace itself such as, for example, in an operator control room.
The objects, features, and advantages Gf the present invention are readily apparent from the following detailed description of the best mode for carrying out the invention when taken in connection with the accompanying drawings.

BRIEF ~ESCRIPTION OF THE ~RAWTNGS
Figure 1 illustrates one type of glass processing ~ystem which might employ the position controller of the present invention~
Figure 2 i8 a bloc~ diagram of the controller of the present invention;
Figure 3 illustrates a general velocity profile utillzed in the pre~ent invention;
Figure 4 i8 a block diagram of the ba~ic functions performed by the slave computer:
Figure 5A i~ a flow chart detailing the basic functions performed by the slave computer;
Figure SB is a continuation of the flow chart of Figure 5A;
Figure 6 i8 a continuation of the flow chart of Figure S, detailing the automatic oscil-late function;

Figure 7 i8 a continuation of the flow chart of Figure 6;
Figure 8 is a flow chart of the opera-tions conducted by the slave computer in formatting a command for a new move;
Figure 9 is a flow chart detailing the velocity calculation a~d selection function;
Figure lO i8 a flow chart illustrating the steps taken in calculating the switchover point for the deceleration ramp;
Figure 11 i~ a flow chart of the zero speed detection feature;
Fiqure 12 is an illustration of another type of glass proce6sing furnace which might employ the position controller of the present invention;
and Figure 13 i8 a partial view of a glass processing system including the optical scanner and fl~g utilized in the automatic thermal expansion compensator~
Figure 14 i8 a top view of a gla~s processing system including the optical scanner and flag utilized in the automatic thermal expansion compen~sator;
Figure 15 is a flow chart of the automat-ic thermal expansion compensation feature;
Figure 16 is a continuation of the flow chart of Figure 15; and Figure 17 illustrateY the calculation of a linear offset according to the routine charted in Figure 10.

BEST MO~E FOR CARRYING OUT THE INVEN~ION

Referring to Figure 1, a glass sheet processinq ~ystem indicated genexally by the reference numeral 10 includes a schematically indicated furnace 12 having a heating chamber 14 within which glass sheet are heated and bent. The system typically includes a roller conveyor 16 including a plurality of rollers 18 that support glass sheets during conveyance into and out of the heating chamber 14.
The gla~s processing svstem 10 also typically includes one or more forming stations 20, 22 and a quench station 24. The forming stations may include one or more curved molds 26-30. The curved molds 26-30 may be of a peripheral ring type having an open center or a continuou~ 6urface mold depending on the particular glass sheet being formed. The curved surface 32 of the mold 26 has a generally concave shape in an upwardly facing directlon such that peripheral portions of the gla88 ~heot lnitially engage the mold and the center of the gla~s sheet thereafter deforms downwardly under the impetus of gravity toward the shape of the mold to initially form the glass sheet.
A curved mold 28 may be employed in con~unction with a second curved mold 30 which can be moved toward~ mold 28 to press the glass sheet therebetween, bending the sheet in conformance with the curved surfaces of the molds 28-30. Additional details of this glass processing system are . ' ` ''.- ., '; ~ ' ;. . " . ..... ' . ' ', ',`': . .' ~' ' '' ' ' ,':

". ' .,. ' ` ' ' ' . ' .' ; '' :' . ' ' ' disclosed in U.S. Patent No. 4,575,390. Another type of processing furnace whlch might incorporate the controller system of the present invention is shown in Figure 12. Other gla~s processing systems employing movable molds and/or other components whlch are required to move to preselected points at variable speeds mlght also employ the controller of the present invention.
A separate, variable speed drive (not ~hown) drlves each of the molds 26-30 about a single axis. Conventional varlable speed drive mechanisms are preferably employed for this purpose.
Flgure 2 illustrates, in block form, the position controller of the present invention which may be utilized to control one or more of the movable molds 26-30 or other like components in a glass processing system. The positioning control system of the present inventlon, generally referred to as 40, includes a slave computer 42, a dlgital to analog converter 56, a variable speed drive 46, and an absolute position encoder 48.
The slave computer 42 includes a microprocessor, static random access memory (RAM), and at least three communication ports, preferably in the form of a serial RS-232-C port 50, for communicating with the master computer, a parallel port 52 for communicatinq with the absolute position encoder 48, and an analog llne 54 connected to the variable ~peed drive 46. A Quitable proyrammed ~MIKUL 6809-4 monocard mlcrocomputer, manufactured * Trade-mark 13 by ~L Industries, Inc., Norwood, Ohio, is prefera-bly employed and includes the components _hown in Figure 2 as the slave computer 42 and the digital to analog converter 44. The MIRUL 6809-4 has a Motorola 6809 microprocessor, serial RS-232-C port, four parallel I/O ports, a real time clock, up to 4K bytex of qtatic RAM, up to 32R byteq of EPROM as well as a ~/A converter.
~he absolute position encoder 48 i8 preferably a 16 bit re~olver of the type commer-cially available from Computer Conver6ions Corpora-tion, Ea~t Northport, New York. However, a conven-tional optical encoder of suitable resolution may be substituted for this purpose.
The slave computer 42 of the position controller 40 is driven by a master computer 44 which is connected for two-way communication by the RS-232-C serial port 50. The ma~ter computer is also preferably a MIRUL 6809-4 monocard microcompu-ter and is suitably programmed to monitor various selected conditions in the glass proces~ing system, such a~ current temperatures and current positions of other movable components (with the aid of information received from the slave computers controlllng these component~). The master computer also serve~ as the receptor of operator input via a suitable operator interface 58 such as a conven-tional data input terminal or other data lnput device.
~hrough direct operator input, or as a result of a preprogrammed action, the ma~ter computer 44 may download a series of move command~

. : . . .: .. .. . . . : :... . : ; ~. ........ ;

.. . . . . . . . . . . . .

or end point positions and velocity profiles for selected end points to the slave computer. The main computer 44 may also interrogate the slave computer 42 for current position information or position and velocity profile data for selected end points via the two-way communication~ link 50 to the slave computer 42.
The slave computer 42 provides the master computer 44 with requested current position infor-mation ascertained from the absolute positionencoder 48, memory and the variable speed drive 46, and performs the necessary calculations to deter-mine and generate a signal which drives the vari-able speed drive 46, thereby moving the associated mold to a preselected position at a preselected velocity profile.
The position controller 40 under the yuidance of the slave computer 42 programmed in a manner as described in fuller detail hereinafter, perform~ the tasks necessary to po~ition the mold 30 within the gla6~ proces~ing system 10, monitor lts mo~ement, and report selected information back to the master computer 44.
This Qeparate position control subsystem improves the modularity of the glass processing control system. Also, features nece8sary and specific to the po~itioning of the mold driven by the position controller 40 are accomplished in the slave computer's 42 system thereby reducing the complexity of the master computer's 44 system.
Increased modularity, coupled with increased communications capability between the slave computer 42 of the position control syste~ 40 with the ma~ter computer 44 in the glass proce~sing system 10 also simplifies debugging, maintenance and modification operations on both the master and slave sy~tems.
In a glas~ processing system 10 of the type shown in Figure 1, it is desirable that the various ~ovable components such as the molds 26-30 can be moved to certain preselected point~ at certain times during the glass tempering/bending process. It is al~o de~irable that the mold be positioned from its present po~ition to the next desired end point at a certain preselected velocity profile. As shown in Figure 3, the typical profile lS includes a controlled, time-based acceleration from the mold's current position Pl until the ~old reaches the lower of either a preselected maximum velocity Vm or the indicated velocity on the deceleratlon curve 12, at which time the mold moves toward its intended end point at this maximum velocity. A~ lt nears the end point, the mold d-celerate~ to a stop, preferably at the desired end point P2. It ~hould be noted that, though the system qenerally ~elects the le~ser of the 11, Vm, 12, or 13 velocities, this profile, and thus the velocity selected, changes considerably with the operator's choice of par~meters relating to these velocity curves. In particular, the operator'~
choice of a maximum velocity V~ may be set at a value that i8 80 high ~8 illustrated by 14) that the system always selects a velocity from 11, 12, or 13 on a programmed motion from P1 to P2.

~. :.. . ...... . .. .. ~

1 . . . ,. . . . ,. , ;, . ... , , , . i . -, . - , . ,, .- .. , ... ~ .... : - .

As will be discussed in further detail below, a set of parameters defining the character-istics of a particular velocity profile for each end point is downloaded from the master computer 44 to the slave computer 42 in the positioning con-troller 40. From these parameters, a unique velocity profile generally of the form shown in Figure 3 i~ generated for the motion to the end point associated with that velocity profile. ~he slave computer then generates the appropriate 6ignal to the variable ~peed drive to achieve the positioning of the mold to the selected end point with the speed changes necessary to match the associated velocity profile.
The set of parameters associated with each particular end point and its velocity profile is downloaded from the master computer 44. These parameters include a statu~ byte ~which may be set to indicate whether thermal expanaion compensation or some other selectable feature i8 de6ired for any motion to thl~ particular end polnt), the identity of an echo back character to be used to tell the master computer 44 when the mold i8 in the request-ed poEition, an acceleration rate (that i8, the rate of increa~e of velocity during the start-up portion of the motion), a deceleration rate (that is, the rate of decrease of velocity during the end portion of the motion), the maximum de~ired veloci-ty for this motion, the linear ramp offset ~that is, the distance from the end poin~ at which the deceleration profile switches from the curve defined as 12 to 13), a delta value representing the change in distance, and a delta value repre-senting the change in velocity which toqether define the slope of the linear ramp, 13, and an encoder 48 value for the end point of this move.
As will be hereinafter described in qreater detail, the slave computer 42 of the position controller 40 is programmed to retrieve these parameters from its RAM whenever it receives a command from the ma~ter computer 44 requesting a move to that end point, and create the requested unique profile for that move.
Figure 4 illu6trates the ba~ic functions of the slave computer 42. Upon receipt of a ~tring of characters from the master computer 44, the slave computer first determines whether that 6tring of characters corresponds to data, such a8 new position and velocity profile data for a selected end point, or a command. The computer then deter-mines whether the command is a request for informa-tion, such a8 the current position of the mold or the current positlon and velocity proflle of a particulsr end polnt, or a command to move the mold to one of the programmed end points.
~he ~lave computer 42 preferably includes means for ~orting the data or COD ands received from the master computer according to a preaelected hierarchy. In the preferred embodiment, motion commands are placed in an input buffer for priority commands, while information requests and downloaded data are placed in a nonpriority input buffer. As priority commands are received and placed in the input buffer, proces~ing of nonpriority commands -and/or data downloading is suspended pending proce~sing of the priority commands. One way of implementing this hierarchical processing is by use of commercially available operating systems having foreground and background processing. Thus, by establishing motion commands as top priority, motion commands, data downloading, and information inquirie6 may be input in any order and at any time without interfering with the operation of the glass ~heet proce~sing sy~tem.
Referring again to Figure 4, if the slave computer 40 receives a motion command, the system next determines whether the mold is currently in the position correspondin~ to the desired end point. If it i8 necessary to move the mold, the system retrieves from memory the velocity profile parameters associated with this particular end point and utilize~ these par~meters to format a velocity profile for this motion. The veloclty profile i8 formatted by adapting a generalized set o~ accoleration curves Rtored in a table in RAM in the ~lave computer 42, and ~dentified by 11, 12, and 13 ln Figure 3 to the particular acceleration parameters programmed for thi~ end point. The 2~ sy~tem then determines the appropriate velocity by choosing the lowest of the start-up acceleration velocity (11), the maximum selected velocity (V~), the deceleration velocity (12), or the linear ramp approach velocity ~13). Once the correct velocity is determined, the system is~ues a signal suffi-cient to activate the variable speed drive at the required velocity. This sequence of steps is ~.:

... :- ..~:

repeated until the current po~ition information, received from the ab601ute poi6ition encoder 48, indicates that the mold i6 in position. At thi6 point the slave computer 42 ~endis an ~echo back~
character to the master computer 44.
Fi~ure~ 5A and 5B illustrate the sy6tem function for processinq a mo~e command in greater detail. The slave computer 42 retrieves the next command from an input buffer in the slave computer 42 which retains a queue of commands received from th~ master computer 44.
If the command is an emergency stop, the sy6tem install~ the current po~ition a6 the new end point, then (proceeding to point A2 in the flow chart) c~lculates the required speed command based upon the current position and sends the appropriate signal to the ~/A converter 4~. ~his 6ignal i8 preferably in the form of ~ 10-bit 61gnal which i8 converted to a corre6ponding voltage signal for tran~mis6ion to the varlable speed drive 46.
If the command is not an emergency stop command, the system follows the 6teps along the path from the point identified at 102. AEi6uming the command is not a PO~ITIVE BUKP or NEGA~IVE BUMP
comm~nd, the system skips the taFkis outlined by the dashed box identified a~ 104 and searches lt~ table of programmed end points and drive profile parame-ters to determine if it can perform the reque6ted move (at 106).
If the requested end point and as~ociated velocity profile i~ found in the table, the 6ystem proceeds to format the velocity profile for this end point (at 108). After the uni~ue velocity profile for that end point is formatted, the accelerating t~me interval variable i~ ~et to zero and the appropriate speed is calculated a~ previ-ously de~cribed (at 110). The accelerating timeinterval variable equals the value of t used to obtain the corresponding velocity from the linear acceleration portion of the velocity profile, 11.
Once the appropriate speed is calculated as more fully described hereinafter ~n connection with Figures 9 and 10, a 10-bit digital signal is ~ent to the ~/A converter 56 (shown ~n Figure 2).
~his 10-bit digital signal i8 then converted by the ~/A converter 56 to a corresponding voltage which is tran~mitted to the var~able speed drive. The 6peed calculation at 110 and subse~uent signal generation is repeated until the current position equals the desired end point.
Referring again to Figure 5A at 102 on the flow chart, if the command i8 a POSITIVE or NEGATIVE ~UMP command, the system gets the encoder ~caler, that i8 the number of units on the absolute position encoder corresponding to a single unit recognized by the operator, and, if the command is a POSI~IVE ~UMP, adjusts the end point by that scaler value. If the command is a NEGATIVE BUMP, it subtracts the ~ame scaler value from the end point.
It should be noted that the software jog function employed by the slave computer i8 prefera-bly implemented as a specific command which, when received by the slave computer system, generate~ a ' .-' '. ' ' . `, .; ` - ', " '~ ' .' ' '. '. '"`, . '.''' '' ' move to a predeined, distant end point. The velocity profile a~ociated with this special end point is typically characterized by a gradual linear acceleration followed by motion at a rela-tively low con~tant velocity Vm. Becau6e the endpoint corresponding to a software jog command is di~tant, the system generate~ ~ignals directing the variable speed drive to move at the constant velocity until the operator releases the jog switch. At this time, an emergency stop causes the driven mold to be stopped and the current positlon to he established ac the de6ired end point.

Formatting Velocitv Profile The function of formatting the velocity profile, shown at 108 on Figure 5B, i8 shown in greater detail in Figure 8.
~or each motion to an end point, the parameters defining the particular velocity profile to that end point mu~t be retrieved from RAM and utillzed ln con~unction with a normalized ~et of profile curves 11, 12, 13 stored in the slave computer's 42 memory to create a velocity profile that is specific to that end point.
The gener~l velocity profile utllized in the present invention, shown in Flgure 3, includes a first portion repre~ented ~y 11 wherein velocity is mea~ured as a function of time. At each point ln time from to ~the time of the start of motion from the current po~ition to the desired end point) the velocity from which the signal is computed increases at a linear rate. By relating desired ., ;, : . . ` ' ' . , , ` . ` 1 ' " , , '' ,. ', ; ~ ', , , ' . . . ' ' ' . ~ ! .

1 32830~

velocity to elap~ed tlme during the initial stages of the motion (11) a controlled, time-based start-up and rate of increase in velocity can be achiev-ed. ~hi~ rate of increase in velocity is, of course, determined by the programmed acceleration rate for the particular end point destination of this motion.
At the point in time the velocity ascer-tained from 11 i~ equal to or greater to Vm, Vm iB
then chosen as the desired velocity and the appro-priate signal for that veloc~ty i8 sent to the variable speed drive. This constant velocity is maintained, irrespective of elapsed time, until the mold reaches a distance X from the end point at which the deceleration ve]ocity repre~ented by 12 for that distance X from the end point is less than or equal to Vm.
In order to achieve greater control over the mold as it approaches the end point, the volocity profile governing the deceleration portion of the move is a function of the distance from the point. At this point the rate of decrease $n velocity follows the path illu~trated by 12 until the mold reacheR its end point.
To provide a more stable deceleration upon the approach to the programmed end point, a linear ramp 13 may be added to the velocity pro-file. Aæ illuætrated in Fiqure 3, this ramp calls for a lower rate of change in velocity per unit distance of the varia~le speed drive during the approach, increasing the likelihood of a 0mooth and accurate finish to the motion.

Figure 8 illustrate~ in greater detail the ~equer,ce of events necessary for formatting a velocity profile, the operation performed at 106 in the general equence of event6 shown in Figure 4.
The sy~tem fir~t determine6 whether a move to the de~ired end point require~ automatic thermal expansion compensation. If 6uch compensation is required, an ATEC flag iR set at thi~ point. Next, the time period acceleration constant is calculat-ed. This con~tant corresponds to the desired rateof change of velocity for each time interval for the initial acceleration curve on the velocity profile (i.e., the slope of 11 in Figure 3~. This acceleration constant is a function of the acceler-ation rate parameter downloaded from the ~astercomputer 44 and an acceleration constant relating to the specific physical characteristics of the variable speed drive for that system.
The deceleration caler is then calculst-ed. ~his caler i8 a6~0ciated with the point in the velocity proflle curve where velocity i8 calculated as a function of distance (denoted by 12 in Figure 3). The profile of this curve is repre-~ented generally by the equation V ~ R ~ ~

Where V is velocity, R i8 a drive parame-ter which again i~ a function of the specific hardware characteri~tic~ of the varia~le peed drive in the system, Ad is the desired deceleration rate programmed for that particular velocity ,. ~.. . . . ,: . . . ........................... . : ~ . .

, . ; . . . . ... - . . .. - . .

profile, and X is the distance between the current position and the desired end point.
Referring again to Pigure 8, the maximum 6peed constant, Vm, for this end point, i8 re-trieved from the parameters downloaded from themaster computer 44.
~ he linear ra~p portion 13 of the veloci-ty profile i~ characteri~ed by two factors. ~he first, the programmed offset, i8 the 0aximum distance from the end point at which the operator wishes the motion to switch from the more rapidly chanqing velocities characteristic of the decelera-tion curve Of 12 to the linear approach ramp of 13.
The second factor i~ the desired slope for 13.
This second factor is represented by the ratio of the delta parameters stored for this end point.
Each of the programmed offset and slope parameter~
are utilized in calculating the specific lin-ear/table offset for this end point. This offset represent6 the amount that the deceleration curve, repre~ented by 12, must be shifted in order to ensure that the velocity indicated on 12 for a point X, corresponding to the programmed offset, i8 identical to the velocity retr$eved from 13 at that point. Additional details relating to this calcu-latlon are described hereinafter in connection with Figures 9 and 10.
Referring again to Pigure 8, calculation of the linear/table offset is followed by a check to see if the automatic thermal expansion compensa-tor flag i~ ~et. If automatic thermal expansion compensation is required for this end point, the adjustment i6 c~lculated, completing the task of formattinq the velocity profile for this end point.
It ~hould be noted that the entire velocity profile can be normalized with specific parameters for each velocity profile associated with a ~pecific end point ~such a6 the slope of 11, the value of Vm, the deceleration rate of 12, and the slope of 13) 80 that the normalized profile can be scaled with a minimum transfer of data from the ma6ter computer 4~ and minimum storage of data in the slave computer's 40 memory.
It will be appreciated by those ~killed in the art that, by storing a normalized profile in memory in the slave computer 42, and creating a specific profile for each programmed end point by scaling that normalized velocity profile with the specific parameters associated with that end point, a relatively large number of end points and associ-ated velocity profile parameters can be stored.
In contra~t, prior position controllers required that end polnt values and the entire table of values representing the associated velocity profiles for each end point ~e stored in ROM in the slave controller. In ~ddition to using a relative-ly large amount of memory, the value~i could not be reprogrammed without removal of the ROM chip from the position controller board.

Velocity Calculation Figure 9 details the velocity calculation feature employed by the positioning system of the present invention and referred to at 110 of Figure 5B. The syste~ first determine~ whether automatic thermal expansion compensation has been selected for thi~ point and performs initial tasks a6~0ciat-ed with that function a6 described hereinafter in S connection with Fiqures 15 and 16. ~he 6ystem then reads the current position from the absolute ~ocition encoder and ~ubtracts this position from the desired end point to obtain Result, equal to the di~tance X from the desired end point. If Result i8 neqative, a flag i6 set indicating that condition. If Result i~ within the linear off~et calculated (as described in Figure 10) the ramp velocity, that is the velocity on the linear deceleration ramp shown as 13, is obtained. If that velocity is greater than Vm, Vm is substituted for 13 velocity. If a zero speed condition is not detected, the velocity is saved and the current position i8 read from the absolute po~ition encod-er. If the ~ystem is not currently in the de~ired end polnt position, the operatlon i8 terminated.
If the system determines that the mold i6 in the desired end point po~ition, the in-position flag i8 set and the in-position character i6 echoed back to the master computer.
Referring to the point indicated a6 200 in Figure 9, if Result is not within the calculated linear offset, the linear off~et is added to Result, and the velocity associated with this value i8 obtained from the deceleration curve, 12. The lesser of this velocity and the velocity obtained from the acceleration velocity of 11 i8 then 6elected and the 6ystem proceeds from point B in the flow chart as previou61y de~cribed.
It ~hould be noted that if a zero speed condition i8 detected, the acceleration velocity, obtained from 11 as a function of the current value of the acceleration time interval variable, i~
saved in place of any previously selected velocity.
This ensures that a controlled, time-based start-up will occur whenever the mold iE detected to be in a ~topped position.
Referring to Figures 10 and 17, the calculation of the linear offset for a particular end point begins with the sy~tem retrieving the current value of the linear offset from the veloci-ty profile format buffer. The desired slope of thelinear ramp, 13, is also obtained. ~he ~ystem then checks to see if the desired offset for this end point is equal to zero. If it is, no linear deceleratlon ramp i8 deslred and the system exits this routine. If a de~ired linear offset i8 programmed for this end point, the velocity on the linear ramp, 13, is determined for X equal to that desired offset. As ~hown in Figure 17, the veloci-ty, Vt, obtained from the linear ramp, 13, at a point X0 equal to the deslred linear offset repre-sents the transition velocity from the deceleration curve f 12 to the linear deceleration ramp of 13.
Referring again to Figure 10, at 300, the transi-tion velocity, Vt, is ~aved and a variable, Dis-tance, 18 set equal to 1. ~i~tance is a variablerepresenting the distance from the desired end point, Xe. ~he velocity for this di~tance i~

-obtained from the deceleration curve, 12, and this velocity is compared to the transition velocity, Vt. If this velocity i6 less than the transition velocity, the value of ~i~tance is incremented by one unit and a new velocity i8 obtained from 12.
~hi~ velocity is again compared to Vt and the prOCe8~ i8 repeated until a velocity, Vd, is obtained from 12 that is equal to Vt. At this point, the value of ~i6tance is equal to the desired linear offset. Thus, any velocities obtained from 12 will be determined as a function of the value of Result, plus the calculated linear offset, as described in Figure 9. It will be appreciated by those skilled in the art that, by adding the linear off~et to Re-cult, the velocity obtained from 12 will decrea~e along a curve shown a6 14 in Figure 17 until, at a distance X0, the transition velocity obtained from the deceleration curve i equal to the transition velocity obtained from the deceleration ramp of 13. Thus, this offset ensures a smooth transition from the decel-eration curve to the more gradu~l deceleration ramp of 13- ;
Zero SPeed ~etection Another feature of the pre~ent invention, zero speed detection, ensures that the start-up acceleration is maintained even though the mold is momentarily stopped.
As shown in Figure 3, the system ascer-tains the desired velocity for each point in time according to the profile established by 11 during the initial stage6 of motion. If the mold i~
~topped for a long enough period of time, the point along 11 at which the velocity for that time i ascertained will yield a velocity in excess of Vm.
While during a normal motion it is desirable to ~witch from this linear rate of accéleration to a constant velocity, if the mold ha6 not moved, it i8 desirable that the system continue to retrieve its velocity for each point in time from 1l to achieve a rapid start-up. The zero speed detection func-tion therefore determines the current position of the mold, by reading the digital input from the absolute position encoder, and compares it with the la~t determined po~ition. If that position i~ un-changed, or has changed by less than a programmablethreshold, and time has elapsed to the extent that the velocity ascertained from 1l on the profile curve is greater than Vm, the possibility of selection of Vm is disabled and the variable speed drive is driven at the linear rate of acceleration defined in 11 with the time re~et to to.
Referring to the flow chart of Figure 11, the zero speed detection routine begins with a determination of whether the mold is in the desired position. If it is, the system exitE from this routine. If not, the current position is read from the encoder, and this position i~ compared to the last position. If the change in po~ition is greater than a programmed threshold value, then the current position is saved a~ the la~t pofiition and the ~ystem exits from the routine.

1 3283~3 If the difference between the current po~ition and the last position i~ not greater than a threshold value then the velocity i~ ascertained from 11 of thf velocity profile for that point in time. If th~s velocity is greater than a threshold (maximum) start-up velocity VM, the motion time, tl, iB set to zero to ensure that the velocity is ascertained from 11 along the velocity profile curve. If the velocity ascertained from 11 i8 not greater than the threshold start-up velocity, a flag iE ~et to en~ure that the acceleration veloci-ty is selected and the system exits. It should be not~ed that forced selection of the linear accelera-tion rate velocity of 11 ensure6 that the velocity signal output to the variable speed drive c2~ses an efficient start-up from the stopped position.

Oscillate Command Function Figures 5A, 6 and 7 detail the cpecific function~ performed by the slave computer 42 in connection wlth an o~cillate command. It is sometlmes desirable to move the molds 26-30 to a selected end point and then oscillate back and forth hetween two points in proximity to that end point, such as during a quench cycle. For this purpo~e, a command essentially in the form of three motion commands strung together, may be sent by the master computer 44 to the slave computer 42. As shown at 106 of ~igure 5A, if a search of the command table results in retrieval of three motion commands side by side (that is, without terminator characters between the command6) the 6ystem pro-ce~es the command as an o~cillate command. ~he first programmed end point is ~aved as a typical S motion command (112), the second programmed end point i~ saved as the ~occillate in point" (114), and the third programmed end point is saved a~ the ~oscillate out po~nt~. The o~cillate flag i6 then set and normal motion processing continues as described a~ove and illustrated in Figure 5B.
If, during normal proces6ing, the 6ystem determines that the driven mold is currently oscillating or that a motion command to oscillate has been issued (shown at 116 on Figure 5B) the oscillate routine of Pigure 7 i8 implemented. The ~ystem determined from the absolute position encoder whether the driven mold is in position. If it is not, the system return~ to norm~l processing.
If it is in position, the system retrieves the current oscillstion command, determines whether it 18 an ~o~cillate in~ or ~oscillate out~ command, echoe~ an in po~ition character to the master computer 44 and returns, at A2, for normal process-ing. If a command to oscillate has just been read by the slave computer 42, the sy6tem enter6 the 06cillate routine at OI. ~he ~ystem then deter-mines whether the driven mold i~ currently in position. If it i~ not, it proceeds with normal motion processing, shown at entry point C on Figure SB. If the driven mold is in position, the system pauses for a time delay, referred to as a German delay, and then sets up the move to the first 1 3283~3 06cillation point, thereby initiating the o~cilla-tion routine. It should be noted that the length of the C.erman delay can be progra~med by the operator.

Automatic Thermal Expansion Compen~ation Referring to Figure 12, some gla~æ pro-cessing sy6tems which employ the position control-ler of the present invention may include a ring mold 200 which i8 movable on a shuttle 206 along a horizontal axis into and out of the heating chamber 202 during the process. Exposure of the shuttle 206 to varying temperature~ during the process may cause thermal expansion of the shuttle 206 and, consequently, a displacement of the centerline of the mold 200 along the axis of motion of the mold 200.
Since it i~ often important that the mold 200 be positioned by the shuttle 206 80 that the centerline 18 at an exact position within the heating chamber 202, for example, during mating w~th complimentary mold 204, this thermal expansion may cause a considerable problem.
~he present system takes advantage of the capability of the slave computer to dynamically alter the location of programmed end points at any time to ad~ust for changes in length of the shuttle due to thermal expansion or contraction.
Referring to Figures 13 and 14, an initial encoder reading is taken when a flag 300, preferably located at the centerline of the mold 200, crosses to a fixed interrogation point 302 on .

:' . ~ ' . . . ~' ' " :;. ' :, ' " , '' ' ' ' .' ' : '."

~ 328303 the machine, preferably an optical scanner, during the initial motion of the shuttle 206. At each subsequent point ln time when the flag 300 pas6es the optical scanner 302, the encoder reading is ascertalned and compared with the initial encoder position.
Any change ln this positlon, presumed to be cau~ed by thermal expanslon or contractlon of the shuttle 206, may then be added or subtracted from the programmed end point so that attempted moves to that end point will result in proper alignment of the mold.
A more detailed description of the automatic thermal expansion compensation is illustrated in Figures 15 and 16. The initial absolute position encoder 48 reading is taken when it ls determined that the optical scanner 302 beam has been interrupted and the absolute position encoder 48 reading for the eye reference i~ equal to zero At this point, the Original Eye Reference is set equal to the current posltlon read from the absolute position encoder 48 ~Figure 2).
Referring to Figure 15, during processing of a move, the system determines whether ATEC is required for that end point. If ATEC is required the system then checks to determine whether the eye reference is greater than zero. If it is, the system retrieves the Original Eye Reference and computes a result, equal to the current eye reference minus the Origlnal Eye Reference.
The ~ystem then determine~ whether the magnltude of this re~ult is greater than a proqrammable maximum reference threshold Rm. If lt i~ not greater than 3g P-42~ -35-this threshold, it i8 as~u~ed that thi~ result is a difference in length due to thermal expan6ion or contraction of the ~huttle 206, and the position of the end point in the control buffer is adjusted by the value of the re6ult. If the result i8 greater than the maximum threshold and this is the first time that the optical scanner ha~ been interrupted, the original reference is ~et equal to the current eye reference. If the result is greater than the thre~hold reference value R~ and this is not the first time the optical ~canner has been interrupt-ed, an error message i8 sent to the master computer at 44.
~his invention has been described in an illustrative manner and it i8 to be understood that the terminology which has been used is intended to be in the nature of words of description rsther than limitation.
Many modifications and variations of the present invention are pos~ible in light of the above teachings. It i8 ~ therefore, to be under-stood that within the scope of the appended claims, the invention may be practiced otherwise than as 6pecifically described.

.. . " , , ............. . , .. ; ,. . . .

. : . . ` '

Claims

1. In a glass sheet processing system, a position control-ler for positioning a movable component driven by a variable speed drive, the position controller being adapted for use with a cen-tral control system which monitors and controls the processing of the glass sheets, the central control system including a master computer having means for providing command signals and data sig-nals relating to preselected end points, the position controller including a position encoder connected to the movable component, and a slave computer, the slave computer comprising:
a first input connected to the master computer for receiving digital signals corresponding to positioning commands, position data and velocity profile data from the master computer;
a second input connected to the position encoder for receiving digital signals corresponding to current position infor-mation from the position encoder;
a first output connected to the variable speed drive for providing control signals to the variable speed drive;
a second output connected to the master computer for transmitting digital signals corresponding to command acknowledgement, error, position data, and velocity profile data to the master computer; and logic means for calculating the distance required to position the movable component from the component's current position to a preselected point, calculating the desired current velocity according to position and velocity profile data for that preselected point, and calculating a digital control signal necessary to drive the variable speed drive to move the movable component to the preselected point at the desired current velocity in response to a positioning command received from the master computer, and transmitting an acknow-ledge signal to the master computer when the movable component has reached the preselected point.

2. The position controller of claim 1 wherein the movable component includes a mold mounted on a shuttle.

3. The position controller of claim 1 wherein the variable speed drive is activated by an analog signal and the position controller further includes a digital to analog converter intercon-nected between the slave computer and the variable speed drive for receiving the digital control signal from the slave computer and converting it to a corresponding analog signal for driving the variable speed drive.

4. The position controller of claim 1 wherein the slave computer includes random access memory for storing the position data and velocity profile data downloaded from the master computer via the first input before or during operation of the controller.

5. The position controller of claim 4 wherein the velocity profile data includes a set of parameters associated with each preselected end point.

6. The position controller of claim 5 further including:
data stored in the memory of the slave computer corresponding to a normalized velocity profile;
logic means for combining the set of parameters associated with a particular end point with the normalized velocity profile to determine a particular velocity profile for that preselected end point; and logic means for periodically selecting the appropriate velocity from the particular velocity profile for output to the variable speed drive during a motion from the movable component's current position to the preselected end point.

7. The position controller of claim 6 wherein the particular velocity profile generated for a preselected end point includes a first portion, 11, wherein velocity is increasing at a generally linear rate as a function of time, a second portion, vm, wherein velocity is constant and a third portion, 12, wherein velocity decrease&

as a function of the current distance of the movable component from the preselected end point.

8. The position controller of claim 7 wherein the third portion 12, of the particular profile generated for a preselected end point is represented by the equation .

9. The position controller of claim 7 wherein the particular velocity profile generated for a preselected endpoint further includes a fourth portion, 13, wherein velocity decreases at a generally linear rate at a function of the current distance of the movable component from the prese-lected endpoint.

10. The position controller of claim 9 further including logic means for calculating a transition velocity, equal to the velocity indica-ted on the fourth portion, 13, of the particular velocity profile for a preselected end point at a preselected linear offset, and logic means for shifting the third portion, 12, of the particular velocity profile so that the velocity obtained from the fourth portion, 13, is equal to the velocity obtained from the third portion, 12 when the current distance of the movable component from the preselected endpoint is equal to the linear offset.

11. The position controller of claim 10 wherein the parameters associated with each prese-lected endpoint include:

data corresponding to the slope of the first portion, 11, data corresponding to the constant velocity vm, data corresponding to the slope of the fourth portion, 13, data corresponding to the linear offset, and data corresponding to the encoder value for the preselected endpoint.

12. The position controller of claim 7 further including logic means for determining when the current position of the movable component remains unchanged for a preselected threshold time period, and, if the current position is unchanged, selecting a velocity from the first portion, 11, of the velocity profile with the elapsed time of the motion reset to zero thereby ensuring a smooth start-up of the movable component.

13. The position controller of claim 1 wherein the slave computer includes logic means for sorting the commands and data received from the master computer according to a predefined hierar-chy, and processing those commands and data in a preselected order corresponding to the predefined hierarchy.

14. The position controller of claim 13 wherein the predefined hierarchy indicates that position commands have priority, with data corre-sponding to preselected end points, velocity profile data, or data inquiries being processed only after all position commands have been pro-cessed.

15. The position controller of claim 1 further including:
a flag fixed to the movable compo-nent;
an interrogator located at a fixed interrogation point in the system and adapted to sense the presence of the flag whenever the flag passes the interrogation point; and wherein the slave computer further includes --an input connected to the interrogator for receiving a signal indicating whether the flag is currently located at the interrogation point, memory for storing an original reference position, and logic means for comparing the value of the current position of the movable component to the original reference position whenever the interrogator indicates that the flag is located at the interrogation point and adjusting the value of the preselected endpoint by the difference between the compared positions, thereby insuring that the movable component reaches the preselected endpoint despite any change in the length of the movable component.

16. The position controller of claim 15 wherein the interrogator is an optical scanner.

17. The position controller of claim 1 wherein the slave computer further includes logic means for determining when an oscillate command has been transmitted from the master computer, and thereafter calculating the distance required to position the movable component from its current position to a first preselected point calculating the desired current velocity according to posi-tion and velocity profile data for that preselected point, and calculating a digital control signal necessary to drive the vari-able speed drive to move the movable component to the first pre-selected point at the desired current velocity, calculating the distance required to position the movable component from the first preselected point to a second preselected point, calculating the desired current velocity according to position and velocity pro-file data for that preselected point, and calculating a digital control signal necessary to drive the variable speed drive to move the movable component to the second preselected point at the desired current velocity, calculating the distance required to position the movable component from the second preselected point to a third preselected point, calculating the desired current velocity according to position and velocity profile data for that preselected point, and calculating a digital control signal neces-sary to drive the variable speed drive to move the movable compo-nent to the third preselected point at the desired velocity pro-file, and transmitting an acknowledge signal to the master com-puter when the movable component has completed the oscillation.

18. The position controller of claim 17 wherein the oscilla-tion command consists of three motion commands, the first motion command indicating the first preselected point, the second motion command indicating the second preselected point, and the third motion command indicating the third preselected point, and each of the motion commands is transmitted from the master computer with-out terminator characters between the commands.

19. In a glass sheet processing system, a position control-ler for positioning a movable component driven by a variable speed drive, the position controller being adapted for use with a cen-tral control system which monitors and controls the processing of the glass sheets, the central control system including a master computer having means for providing command signals and data sig-nals relating to preselected end points, the position controller including a position encoder connected to the movable component, and a slave computer, the slave computer comprising:
a first input connected to the master computer for receiving digital signals corresponding to positioning commands, position data and velocity profile data from the master computer;
a second input connected to the position encoder for receiving digital signals corresponding to current position infor-mation from the position encoder;
a first output connected to the variable speed drive for providing control signals to the variable speed drive;
a second output connected to the master computer for transmitting digital signals corresponding to command acknowledg-ment, error, position data, and velocity profile data to the master computer;
random access memory for storing the position data and velocity profile data downloaded from the master computer via the first input before or during operation of the controller;

logic means for calculating the distance required to position the movable component from the component's current posi-tion to a preselected point, calculating the desired current velocity according to position and velocity profile data for that preselected point, and calculating a digital control signal neces-sary to drive the variable speed drive to move the movable compo-nent to the preselected point at the desired current velocity in response to a positioning command received from the master com-puter;
logic means for determining during the positioning of the movable component to the preselected point whether the current position of the movable component remains unchanged for a prese-lected threshold time period, and, if the current position is unchanged, selecting a preselected start-up velocity, rather than the velocity indicated from the velocity profile data, and reset-ting the elapsed time of the motion to zero, thereby insuring a smooth start-up of the movable component;
logic means for transmitting an acknowledge signal to the master computer when the movable component has reached the preselected point.

20. In a glass sheet processing system, a position control-ler for positioning a movable component driven by a variable speed drive, the position controller being adapted for use with a cen-tral control system which monitors and controls the processing of the glass sheets, the central control system including a master computer having means for providing command signals and data sig-nals relating to preselected end points, the position controller including a position encoder connected to the movable component, and a slave computer, the slave computer comprising:

a first input connected to the master computer for receiving digital signals correspon-ding to positioning commands, position data and velocity profile data from the master computer;
a second input connected to the position encoder for receiving digital signals corresponding to current position information from the position encoder;
a first output connected to the variable speed drive for providing control signals to the variable speed drive;
a second output connected to the master computer for transmitting digital signals corres-ponding to command acknowledgment, error, position data, and velocity profile data to the master computer;
logic means for calculating the distance required to position the movable component from the component's current position to a preselected point, calculating the desired current velocity according to position and velocity profile data for that preselected point, and calculating a digital control signal necessary to drive the variable speed drive to move the movable component to the preselected point at the desired current velocity in response to a positioning command received from the master computer;
logic means for determining when the positioning command transmitted from the master computer is an oscillate command, and thereafter:
a) calculating the distance re-quired to position the movable component from its current position to a first preselected point, calculating the desired current velocity according to position and velocity profile data for that preselected point, and calculating a digital control signal necessary to drive the variable speed drive to move the movable component to the first preselected point at the desired current velocity, b) calculating the distance re-quired to position the movable component from the first preselected point to a second preselected point, calculating the desired current velocity according to position and velocity profile data for that preselected point, and calculating a digital control signal necessary to drive the variable speed drive to move the movable component to the second preselected point at the desired current velocity, and c) calculating the distance re-quired to position the movable component from the second preselected point to a third preselected point, calculating the desired current velocity according to position and velocity profile data for that preselected point, and calculating a digital control signal necessary to drive the variable speed drive to move the movable component to the third preselected point at the desired velocity profile.

21. The position controller of claim 20 wherein the slave computer further includes:

memory for storing data corresponding to a preselected time delay; and logic means for determining when the movable component has reached the first preselected point in an oscillation, and thereafter instituting a delay in the motion of the movable component equal to the preselected time delay before gener-ating the control signal necessary to move the com-ponent to the second preselected point.

22. The position controller of claim 20 wherein the first preselected point is equal to the second preselected point.